DESCRIPTION: Investigations of two novel cyclopentanoid natural products will be continued. The first is the antibiotic aristeromycin, a cyclopentyl analog of adenosine produced by Streptomyces citricolor, along with the cyclopentenyl adenosine analog neplanocin A, which is an antitumor and antiviral agent. Previous investigations of aristeromycin biosynthesis established that the cyclopentane ring is generated by C-C bond formation between C-2 and C-6 of glucose, and provided evidence that carbocyclic analogs of the "normal" purine biosynthetic pathway may be intermediates. Also, a novel enzyme that catalyzes the anti-Markovinikov reduction of neplanocin A to aristeromycin was isolated and partially purified from S. citricolor. Future studies will focus on furthe purification of this enzyme, cloning of the gene coding for the enzyme, overexpression of the enzyme, and elucidation of the mechanism of the reductio process. Two unsaturated carbocyclic PRPP analogs that may be intermediates in aristeromycin biosynthesis will also be synthesized and evaluated using cell-free extracts of S. citricolor. The second cyclopentanoid to be investigated is the phytotoxin coronatine, produced by Pseudomonas syringae. The cyclopropyl amino acid moiety of the toxin, coronamic acid, has been found to be biosynthesized from L-alloisoleucine by a highly unusual process that proceeds with retention of the nitrogen atom of the amino acid and the loss of only two hydrogen atoms, one from the alpha-position of the amino acid and one from the methyl group. Future investigations will examine the details of coronamic acid biosynthesis and will be carried out in collaboration with Dr. Bender of Oklahoma State University who has sequenced three genes involved in coronamic acid biosynthesis. The two genes whose role appears clear will be overexpressed and their biochemistry studied. The product of the first gene (CmaA) appears to activate L-alloisoleucine as an acyl adenylate, convert it t an enzyme bound thiol ester, and then clize it to enzyme-bound coronamic acid. A motif in CmaA suggests it is likely to be an iron-dependent dioxygenase. The product of the second gene (CmaT) appears to be a thioesterase whose role is t release coronamic acid from CmaA. Mechanistic studies with both of these enzymes are proposed. The potential use of the CmaA gene for construction of hybrid peptide antibiotics containing L-alloisoleucine or coronamic acid will also be explored, using surfactin biosynthesis in B. subtilis as a model system.